Basal cell carcinoma (BCC), the most common cancer in North America, is thought to be universally driven by dysregulation of the Hedgehog (Hh) signaling pathway. Small molecule inhibitors of Hh, such as vismodegib, have proven to be highly effective at regressing advanced BCCs; however, these drugs do not cure patients of disease, as tumors typically recur at their original sites upon cessation of drug treatment. These observations suggest that most BCCs contain at least a subset of persistent cells that can enter a dormant state upon Hh blockade. To better understand how tumors respond to drug therapy, we have generated a mouse model of BCC whereby deletion of Ptch1 is targeted to hair follicle stem cells. These mice develop nodular BCC-like lesions that can be partially regressed by vismodegib, but recur upon subsequent drug withdrawal, similar to human tumors. In Specific Aim 1 of this proposal, we will refine both in vivo and in vitro tumor regression assays incorporating vismodegib treatment, and characterize residual tumor cells that persist during therapy. In Specific Aim 2, we will perform genetic gain- and loss-of-function studies, coupled with lineage tracing, to ascertain whether manipulating the Notch pathway, a major mediator of differentiation in the skin, can affect vismodegib-induced tumor regression. Findings from these studies will help us better understand the cellular pathways that underlie tumor response to anti-Hh blockade, and may lead to novel therapeutic approaches whereby multiple tumor susceptibilities can be simultaneously targeted to prevent recurrence. Our Notch experiments may also serve as a proof-of-principle study on the potential effectiveness of tumor differentiation therapy.